Rhinos, horses and paraceratheres: untangling a tradition

While searching
for a cladogram that includes rhinos, horses AND paraceratheres, I ran across the following papers and taxa:

Bai et al. 2020 discussed the origin of Rhinoceratoidea and the phylogeny of Ceratomorpha (rhinos). The authors included rhinos and paraceratheres, but not horses (= Equus and Miohippus). An included taxon was Orientolophus (known from a mandible and teeth). The Bai et al. cladogram (their Fig. 7) included several dozen taxa, including tapirs, Heptadon, Hyrachus and many other rhinos, Fostercooperia, Pappaceras, Juxia, Paraceratherium, , but, alas… no horses.

Figure 4. A variety of horse and paracerathere skulls.
Figure 4. A variety of horse and paracerathere skulls.

Bai et al. 2018 wrote:
“Here we report the earliest Eocene Asian equid, Erihippus tingae gen. et sp.
nov., based on partial specimens initially assigned to the ceratomorph Orientolophus hengdongensis, from the Hengyang Basin of Hunan Province, China.”

Initially a rhino. Now a horse. Interesting… so there’s hope.

Bai et al. 2018 continued:
“The nearly simultaneous appearance of equids, ceratomorphs, ancylopods, and brontotheres in the Hengyang Basin suggests that the four main groups of perissodactyls diverged as early as, or no later than, the beginning of the Eocene (about 56 Ma), and displayed different dispersal scenarios during the early Eocene.”

Personal reminder: Ancylopods are chalicotheres and kin.

Bai et al. 2018 continued:
“Based on ancestral reconstructions of the geographic distribution using the parsimonious criterion, we suggest that equids originated in Europe. One clade dispersed to the Indian-subcontinent giving rise to Ghazijhippus, probably along the northern margin of the Neotethys, while the other clade gave rise to European Cymbalophus and Pliolophus quesnoyensis.
The European equids dispersed to North America via the Greenland land bridge giving rise to Arenahippus and Sifrhippus, which immigrated to Asia via the Bering Strait giving rise
to Erihippus during the PETM.”

Much of the data in Bai, Wang and Meng 2018 are dental.
Sadly, their cladogram includes several other horse-like ‘hippus’ taxa, but not Miohippus and Equus and no paraceratheres (Fig. 1). Their outgroup, Radinskya, nests in the large reptile tree (LRT, 1879+ taxa) with Procavia, the hyrax, basal to elephants, not perissodactyls. Their other outgroup taxon, Cambaytherium, nests in the LRT with Cornwallius between hippos and anthracobunids, not close to perissodactyls. Matching the LRT, brontotheres are close to rhinos, but the authors nest Danjiangia with brontotheres. In the LRT Danjiangia is a pig. No pigs are included in the Bai, Wang and Meng 2018 taxon list, so taxon exclusion bites the hand again.

Bottom line:
Try not to cherry-pick taxa. Let your wide-gamut cladogram pick taxa for you. To make sure you have a wide-gamut cladogram, keep adding taxa. Dental traits can converge. Don’t trust genes in deep time studies.

Bai B, Wang Y-Q and Meng J 2018. The divergence and dispersal of early perissodactyls as evidenced by early Eocene equids from Asia. Nature Communications Biology 1:115 | DOI: 10.1038/s42003-018-0116-5 | http://www.nature.com/commsbio
Bai B et al. (4 co-authors) 2020. The origin of Rhinocerotoidea and phylogeny of Ceratomorpha (Mammalia, Perissodactyla). Nature Communications Biology 3:509 | https://doi.org/10.1038/s42003-020-01205-8 | http://www.nature.com/commsbio

Seals, sea lions and a ‘possible parallel evolution’ according to Patterson et al. 2020

Patterson et al. 2020 combine
morphological and molecular datasets to find, “many traits shared between extant pinnipeds have arisen independently in the three [extant] clades. Thus, while the present
analysis identifies a new suite of morphological synapomorphies for Pinnipedia, the
frequency of reversals and other homoplasies within the clade limit their diagnostic value.”

The three traditional clades of Pinnipeds
comprise the extant families Odobenidae (the walrus), Otariidae (the eared seals: sea lions and fur seals), and Phocidae (the earless seals, or true seals).

taxon exclusion mars Patterson et al. Simply by adding taxa, as in the large reptile tree (LRT, 1882+ taxa; subset Fig. 1) the Pinnipedia is not a monophyletic clade (Fig. 1). The LRT does find parallel evolution in seals + walruses and sea lions, but their last common ancestor is not a ‘pinniped’. Rather their last common ancestor in the LRT was close to Mustela, the European mink.

Figure 1. Subset of the LRT focusing on the placental mammal clade Carnivora documenting the origin of seals (Phoca and kin) arising from sea otters (Enhydra and kin) and a separate origin for sea lions (Zalophus and kin) from Miacis.

Enhydra, the sea otter, was included
in the Patterson et al. cladogram using molecular data, but it nested as a taxon with no descendants.

the wolf, Canis lupus, was specified (= chosen) as the outgroup. In the LRT (Fig. 1) Canis is not basal to seals, sea lions, walruses or even minks. Rather, Canis is a highly derived taxon in the LRT. So, that’s another problem. It would have been better if Patterson et al. had chosen the raccoon, mongoose, kinkajou or coatimundi, even the panda, all basal members of Carnivora, for their outgroup.

Sea lion ancestors in the LRT,
Miacis and Hyopsodus were omitted from the Patterson et al. study.

Seal ancestors in the LRT,
Megistotherium and Palaeosinopa, were also omitted from the Patterson et al. study.

Patterson et al. report
“The Pinnipedimorpha was erected by Berta (1991) and includes Enaliarctos and all of its descendants, including a monophyletic crown-group Pinnipedia, which is strongly supported in all phylogenetic analyses performed in the present study.”

Here in the LRT
the Pinnipeda was shown to be not monophyletic back in 2017. So you heard it here first. If someone suggested this hypothesis of interrelatioships earlier, please provide that citation so I can promote it here. Add taxa to your Carnivora study to confirm or refute the present hypothesis of interrelationships.

Paterson RS, Rybczynski N, Kohno N and Maddin HC 2020. A Total Evidence Phylogenetic Analysis of Pinniped Phylogeny and the Possibility of Parallel Evolution Within a Monophyletic Framework. Front. Ecol. Evol. 7:457. doi: 10.3389/fevo.2019.00457


Gavialis, the gharial, enters the LRT as an overlooked ‘living fossil’

The extant gharial (= gavial) is obviously different
from the other extant crocodiles, alligators and caimans. That’s because it IS phylogenetically different, more different than professors have thought and undergrads have been taught.

According to the LRT
(Fig. 6), the Thalattosuchia, the marine croc clade of the Early Jurassic to Early Cretaceous, now has a living representative, the gharial, Gavialis.

Thalattosuchia are traditional members
of the former paraphyletic clade Mesosuchia and more recently renamed, Mesoeucrocodylia, To be considered monophyletic a clade must include all descendants: in this case, extant crocs and their closest extinct kin (= Eusuchia). There is a short list of traits that differentiate the mesosuschs from the eususchs. In eusuchians the choanae are completely surrounded by pterygoid bones.

Gavialis has that trait,
but did it evolve in parallel from a teleosaur – thalattosuchian ancestor over the last 150 million years? The LRT says so. We don’t want to “Pull a Larry Martin” now, relying on one or a few traits to define a clade. Rather, we want to figure out a last common ancestor using all the traits in the LRT from snout to tail tip and let convergence happen, even if it upsets an old order.

Figure 1. Gavialis skeleton.

According to Wikipedia
(citing Brochu 1997), “The evolution of the gharial and its relationship with and divergence from other crocodilians have been a subject of controversy.”

The false gharial,
(Tomistoma shlegelii, Fig. 5), a ‘genetic’ sister, also enters the large reptile tree (LRT, 1882 taxa; Fig. 6), but not with Gavialis (Figs. 1, 2). Not surprisingly, Tomistoma enters the LRT alongside a very similar Crocodylus (Fig. 4).

Figure 2. Gavialis (gharial) skull from Digimorph.org and used with permission. Colors added here. Compare to Pelagosaurus in figure 3.

Gavialas gangeticus (originally Lacerta gangetica Gmelin 1789; Figs. 1, 2) nests with Late Jurassic Pelagosaurus (Fig. 3), a basal teleosaurid, thalattosuchian taxon with legs (not flippers).

Thalattosuchian outgroups in the LRT include
Dyoplax from the Late Triassic. According to Wikipedia, “Oscar Fraas, the original describer of Dyoplax, described the specimen as having “the head of a lizard and the armor of a gavial”.

Figure 3. Pelagosaurus nests as a sister to Gavialis (Fig. 2) in the LRT.

According to Wikipedia,
in molecular studies Gavialis and Tomistoma are sister taxa within Gavialidae, and sisters to the clade of crocodiles. In those same molecular studies caimans and alligators arise from a nearby branch.

Even when correct, please ignore deep time molecular studies like this.
Too often they deliver false positives along with true positives. Likewise please ignore ‘total evidence‘ studies because these too often include molecular data. Instead just use traits and fossils to find interrelationships, as in the LRT where sister taxa look more alike overall and in detail. It’s a simple if time-consuming task, you can do alone (you don’t need a chem lab and dozens of more or less willing students), and it works.

Figure 4. Skull of Crocodylus. Compare to Tomistoma in figure 5.

According to Wikipedia,
Teleosauridae is a family of extinct typically marine crocodylomorphs similar to the modern gharial that lived during the Jurassic period. Teleosaurids were thalattosuchians closely related to the fully aquatic metriorhynchoids, but were less adapted to an open-ocean, pelagic lifestyle.”

Figure 5. Skull of Tomistoma, the false gharial. Compare to Crocodylus in figure 4.
Figure 6. The LRT nests gharilals at the base of the marine crocs along side Pelagosaurus. This may be a novel hypothesis.

Here’s a YouTube video of croc expert,
Dr. Chris Brochu (2013), running through a long list of interesting “surfboards with eyes” as he lectures on his favorite clade. Dr. Brochu discusses gharials at 10:40.

The nesting of gharials at the base of the Thalattosuchia (= marine crocs)
appears to be a novel hypothesis of interrelationships. If there is an earlier citation, please let me know so I can promote it here.

Pierce et al. 2017 came close
when they referred to basal thalattosuchians as,“‘gavial-like’ near shore predators.” Their figure one (Fig. 7) compares µCT scans of Pelagosaurus and Gavialis. Pierce et al. report, “We chose Gavialis because out of all extant crocodylian species its skull morphology (longirostrine) and ecology (aquatic, piscivorus) are most analogous to that of Pelagosaurus.”

In this case, analogy turned out to be homology. Good choice!

Figure 7. From Pierce et al. 2017 compariing Pelagosaurus to Gavialis. In this case, analogy turned out to be homology.

Cau 2019 did not come close.
In a cladogram focusing on crocodylomorphs, Cau chose an unrelated taxon without descendants, Postosuchus, as an outgroup taxon for the Crocodylomorpha. That flipped the more derived Hesperosuchus with the more basal Junggarsuchus at the base of the Crocodylomorpha. Gracilisuchus, Scleromochlus, Saltopus and other basal bipedal crocs were omitted. Cau nested Gavialis at the base of Alligator + Crocodylus far from the base of Thalattosuchia and Pelagosaurus. Dyoplax was omitted. Cau 2019 did not create his own analysis, but borrowed one from Ösi et al. 2018 [describing Magyarosuchus fitosi, a sister to Pelagosaurus]. That study was based on Young et al. 2016, a study without a cladogram, but a great quote for today’s post, “But, were teleosaurids really just ‘marine gavials’?”

Apparently yes. Why did no one else in the last several decades seem to notice?

Perhaps this is due
to the small, but inevitable evolutionary changes in Gavialis since the final appearance of traditional teleosauridae (Geoffery 1831) about 150 million years ago. It’s time to look at some transitional palates.

According to Wikipedia, “Teleosaurids were originally regarded as marine analogues to modern gharials, as they both typically share long, tubular snouts and narrow teeth.” No we can theink of teleosaurids as marine homologs to modern gharials.

So, that’s why gharials look different
than the rest of the extant crocs. They ARE phylogenetically different in the LRT. They are living representatives of the Thalattosuchia -Teleosauridae, clades traditionally thought to have been extinct for about 150 million years due to everyone “Pulling a Larry Martin”.

Add taxa to your own analysis
and see what you get. In science many others have to test and confirm an hypothesis before it becomes ‘widely accepted’. From my experience, and that of Dr. John Ostrom, this confirmation will take decades.

A few days later, July 11, 2021, three big crocs: Dyrosaurus, Sarcosuchus and Gryphosuchus entered the LRT with no shifts in the tree topology.

Brochu CA 1997. Morphology, fossils, divergence timing, and the phylogenetic relationships of Gavialis. Systematic Biology 46(3):479–522.
Cau A 2019. A revision of the diagnosis and affinities of the metriorhynchoids (Crocodylomorpha, Thalattosuchia) from the Rosso Ammonitico Veronese Formation (Jurassic of Italy) using specimen-level analyses. PeerJ DOI 10.7717/peerj.7364
Gmelin JF 1789. Amphibia – Pisces. In: Gmelin JF (Ed) Caroli a Linnaei Systema Naturae per Regna Tria Naturae, Ed. 13. Tome 1(3). G.E. Beer, Lipsiae [Leipzig]. pp. 1033-1516.
Ösi A, Young MT, Galácz A, Rabi M 2018. A new large-bodied thalattosuchian
crocodyliform from the Lower Jurassic (Toarcian) of Hungary, with further evidence
of the mosaic acquisition of marine adaptations in Metriorhynchoidea. PeerJ
6:e4668 DOI 10.7717/peerj.4668.
Pierce SE, Williams M and Benson RBJ 2017. Virtual reconstruction of the endocranial anatomy of the early Jurassic marine crocodylomorph Pelagosaurus typus (Thalattosuchia). PeerJ. 2017; 5: e3225.
Young MT, Rabi M, Bell MA, Steel L, Foffa D, Sachs S and Peyer K 2016. Big-headed marine crocodyliforms, and why we must be cautious when using extant species as body length proxies for long extinct relatives. Palaeontologia Electronica 19.3.30A:1–14.


The following short video
reports the false gharial is “more related to crocodiles.”

The following longer video
reports the false gharial is a member of the gavialidae. Toward the end, in a desperate effort to show a skeleton, the videographers substituted a Champsosaurus in two view. Yow!

Another gharial video (46 minutes) with host Nick Baker (2018),
IMHO he’s being groomed and introduced to be the next David Attenborough. Good choice!

Beak clapping in shoebill storks and Pteranodon

After building a full scale Pteranodon skeleton (Fig. 1) with a hinged jaw,
(based on the Triebold specimen) several decades ago, I noticed the possibility of some sort of loud clapping noise that could be made by snapping the jaws shut over and over, like clapping hands. This was made possible by the tight fit of the jaw lines amplified by the hollow elements.

Pteranodon model based on the Triebold specimen by David Peters
Figure 8. Pteranodon model based on the Triebold specimen, NMC41-358.

Lo and behold, this very short and delightful YouTube video of a shoebill stork
(genus: Balaeniceps) exhibits the same behavior, this time with a soundtrack. The headline says, “it sounds like a gun fight.” And it does!

Let your imagination take you back
to the Late Cretaceous shores of Hays, Kansas, where flocks of Pteranodon (Fig. 1) were advertising their availability, saying ‘good morning’ or bitching to their neighbors in much the same fashion.

Ptolemaia: another century-long enigma finds a home in the LRT

Miller et al. 2015 concluded:
“Osborn’s [1908] lament that these animals were so peculiar that nothing much could be said about their “habitats or affinities” still stands.”

Miller et al. 2015 wrote:
“Among the Fayumian elements preserved at Nakwai are three specimens attributed to Ptolemaia cf. grangeri (Mammalia: Ptolemaiida). Ptolemaiids are an enigmatic order of mammals best known from Oligocene deposits in the Fayum, Egypt. Most of this material consists of isolated teeth, but all species are represented by at least some gnathic material, and the P. grangeri hypodigm includes one crushed but otherwise fairly complete skull.”

Based on the narrow skull of LRT sister taxa
(Fig. 2) the skull of Ptolemaia (Fig. 1) is not that crushed. Turns out these taxa have a fairly narrow skull morphology.

Figure 1. Ptolemaia grangeri from xxxx about full scale on a 72dpi monitor. The mandible is Ptolemaia lyonsi Osrborn, AMNH 13269.

Ptolemaia grangeri
(P. lyonsi, Osborn 1908; Bown and Simons 1987, 1995; Miller et al. 2015; Oligocene) is a traditional enigma taxon, but nests here in the large reptile tree (LRT, 1881+ taxa; subset Fig. 3) alongside the IVPP V5235 specimen assigned to Hapalodectes. Ptolemaia shows the extent of the premaxilla missing in the IVPP specimen.

Figure 1. Two Hapalodectes specimens. The smaller one nests at the base of the Primates. The larger one nests as the base of the anagalid-tenrec-odontocete clade.
Figure 2. Two Hapalodectes specimens. The smaller one nests at the base of the Primates. The larger one nests as the base of the anagalid-tenrec-odontocete clade.

The IVPP specimen assigned to Hapalodectes
(Fig. 2) was not yet published when most Ptolemaia workers published, but Miller et al. 2015 did not mention it. In the LRT (Fig. 3) these taxa are basal to the clade that includes Anagale, the elephant shrew, Rhynchocyon, the giant elephant shrew, Andrewsarchus, tenrecs, pakicetids, archaeocetids and odontocetes. So all of these taxa might be considered anagalids, but the naming of the clade Ptolemaiidae is decades older. Zalambdalestes is a questionable traditional member of the Anagaloidea, but nests elsewhere (in Glires) in the LRT.

Osborn 1908 created the family Ptolemaiidae
for Ptolemaia lyonsi, the mandible (Fig. 1) he considered, “This problematic form…”

Figure 3. Subset of the LRT focusing on basal placentals and especially the elephant shrew – tenrec – odontocete clade first recognized in the LRT. Now Ptolemaia nests at a basal node, derived from tree shrews more or less resembling the genus Vulpavus (Fig. 4).

Osborn’s 1908 thoughts on the affinities of Ptolemaia:
“This evidently represents a new family of mammals, to which the name Ptolemaiidd may be given. It possibly represents a new order of mammals which will be defined by the writer if additional materials are found. It is obviously not a primate.

“The large size of the coronoid, depressed position of the condyle, subtrenchant characters of the premolars, enlarged cutting teeth tnd tuberculo-sectorial molar teeth, relate it rather to the unguiculate than to the ungulate division of placentals. The elevated trigonid and depressed talonid belong to a primitive stage characteristic of insectivores and creodonts; the dentition does not resemble, however, that of any known form of Unguiculate, either Insectivore, Carnivore, Creodont, or Edentate.

The laterally compressed premolars with cuspules all in the same line suggest those of the Pinnipedia; but the dental formula is entirely different from that of the Pinnipedia. Certain Creodonts have a series of homologous cusps on the premolar teeth but in these carnivores the main cusp exhibits a piercing character which is lacking in this type.

While the anterior teeth of Ptolemaia may have been adapted to the prehension of an active prey, the premolar and molar teeth are not in the least of a carnivorous or sectorial character.”

Osborn was certainly keeping his options open, but he was “Pulling a Larry Martin“, decades prior to software cladograms. So, Osborn was doing the best he could for his time.

Unfortunately for Osborn,
closely comparable fossils (Fig. 2) were described later. Having a growing catalog of taxa in the LRT makes comparisons easier than in Osborn’s day or Miller’s day. Currently there appears to be no need for a separate order ‘Ptolemaiidae’. Ptolemaia tucks in quiet nicely with other, more completely known taxa. Unfortunately this phenomic clade is still novel, not widely known and still not tested. Hopefully someone will someday test the LRT with a competing study using a similar taxon list.

Figure 7. Mink-like Vulpavus (Eocene) is the sister to mink-like Caluromys in the LRT. The larger Vulpavus has one fewer molar, a carnassial lower molar, a narrower zygoma, but otherwise similar traits.
Figure 4. Mink-like Vulpavus (Eocene) is the sister to mink-like Caluromys in the LRT. The larger Vulpavus has one fewer molar, a carnassial lower molar, a narrower zygoma, but otherwise similar traits.

Vulpavus palustris
(Fig. 4) is a basal placental providing clues to the post-crania of Ptolemaia, based on phylogenetic bracketing.

Bown TM and Simons EL 1987. New Oligocene Ptolemaiidae (Mammalia: ?Pantolesta) from the Jebel Qatrani Formation, Fayum Depression, Egypt. Journal of Vertebrate Paleontology. 7 (3): 311–324.
Miller ER, et al. (5 co-authors) 2015. Ptolemaia from West Turkana, Kenya. Bulletin of the Peabody Museum of Natural History. 56 (1): 81–88.
Osborn HF 1908. New fossil mammals from the Fayum Oligocene, Egypt. Bulletin of the American Museum of Natural History. 24: 265–272.
Simons EL and Bown TM 1995. Ptolemaiida, a new order of Mammalia–with description of the cranium of Ptolemaia grangeri. Proceedings of the National Academy of Sciences USA. 92 (8): 3269–73.

Avila and Mothé 2021 try to show how South American wombats (marsupials) arose from African hyraxes (placentals)

Once again,
paleontologists are excluding taxa, testing genes and recovering untenable results.

Unfortunately Avilla and Mothé 2021 used genetic studies
to show South American marsupial wombats, like Pyrotherium (Fig. 2), arose from African placental hyraxes (Procavia and kin, Fig. 3). Astrapotheres (Fig. 4) were also added to this recipe of unrelated taxa.

the authors cherry-picked a short list of taxa following current university textbooks and lectures. That means they excluded a raft of pertinent taxa that would have separated these clades.

From the Avila and Mothé 2021 abstract:
“The South American native ungulates (SANUs) are usually overlooked in Eutherian phylogenetic studies. In the rare studies where they were included, the diversity of SANUs was underrated, keeping their evolutionary history poorly known. Some authors recognized the SANUs as a monophyletic lineage and formally named it Meridiungulata.”

According to Wikipedia
Meridiungulata is an extinct clade with the rank of cohort or superorder, containing the South American ungulates Pyrotheria (possibly including Xenungulata), Astrapotheria, Notoungulata and Litopterna. It is not known if it is a natural group; it is known that both Litopterna and Notoungulata form a clade based on collagen evidence, but the placement of the other members is uncertain. it was erected to distinguish the ungulates of South America from other ungulates.”

Earlier the LRT split up members of the traditional and now defunct clade, Notoungulata. It split apart the pyrotheres from the astrapodtheres and split the litopterns as well.

Figure 1. Cladogram from Avila and Mothé 2021. Color overlays added here. Some fragmentary taxa based on dentition (the clade of red branches) I’m not going to guesstimate.

According to a wide gamut trait-based phylogenetic analysis,
the large reptile tree (LRT,1880+ taxa), members of the Pyrotheria (Fig. 2) are marsupial diprotodont wombats, not placentals. Members of the Liptoterna, like Macrauchenia, are chalicothere sisters (close to living perissodactyls). Members of the Astrapotheria are related to phenacodontids (basal hoofed placentals) and Titanoides.

So if Avila and Mothé are linking these unrelated taxa together
based on DNA, we have one more example of continental viruses infecting and skewing phylogenetic analyses (e.g. Afrotheria, Laurasiatheria, etc).

Figure 3. Pyrotherium is a marsupial, not a relative to Notostylops, contra Billet 2010.
Figure 2. Pyrotherium is a diprotodont (wombat) marsupial in the LRT.

Continuing from the Avila and Mothé 2021 abstract:
“Here, we recognized and defined a new supraordinal lineage of Eutheria, the Sudamericungulata, after performing morphological phylogenetic analyses including all lineages of SANUs and Eutheria. The SANUs resulted as non-monophyletic; thus, Meridiungulata is not a natural group; Litopterna and “Didolodontidae” are Panameriungulata and closer to Laurasiatheria than to other “Meridiungulata” (Astrapotheria, Notoungulata, Pyrotheria, and Xenungulata).”

No Eutheria = Placentalia, so wombats (= Pyrotherium, Fig. 2) are not members.
Yes Liptoterns are perissodactyl placentals, so not closely related to pyrothere marsupials.
Mixed Some traditional notoungulates are placentals in the LRT, others are marsupials. This is why it is a good idea to avoid suprageneric statements based on defunct clades, if possible.

Figure 3. Skeleton of a hyrax (Provavia).
Figure 3. Skeleton of a hyrax (Provavia). Yes, it looks kind of like a wombat, but it is not one.

Continuing from the Avila and Mothé 2021 abstract:
“The other “Meridiungulata” is grouped in the Sudamericungulata, as a new monophyletic lineage of Afrotheria Paenungulata, and shared a common ancestor with Hyracoidea.”

This is false. See the LRT for hyrax ancestors and descendants, none of which are marsupials. This is why we all need to minimize taxon exclusion and avoid deep time genomic studies. Too often DNA studies lead to messes like this, promoting geographic viruses over physical traits.

Figure 4. Astrapotherium to scale with two specimens of Meniscotherium.
Figure 4. Astrapotherium to scale with two specimens of Meniscotherium.

Continuing from the Avila and Mothé 2021 abstract:
“The divergence between the African and South American lineages is estimated to Early Paleocene, and their interrelationships support the Atlantogea (Ezcurra and Agnolin 2012) biogeographic model.”

Let’s not forget that there was a post-Cretaceous faunal exchange between three linked southern continents: South America, Antarctica and Australia. At the same time there was a growing Atlantic Ocean between South America and Africa. The Atlantogea hypothesis (Ezcurra and Agnolin 2012) is not supported by the LRT.

All deep time genomic studies (so far) have been wastes of time and effort.
By that I mean, the resulting sister taxa too often don’t look anything like one another. And they should resemble one another if they are related to one another. Endemic viruses affect DNA testing and split clades into geographic areas, too often gathering dissimilar local taxa together. Just look at the Avila and Mothé 2021 results linking rodents with ungulates and hyraxes with wombats.

Colleagues, please put away your test tubes full of genes
and start building a wide-gamut, trait-based phylogeny, like the LRT. Then you will have the powerful tool you’ve been looking for, one you can use for the rest of your life to avoid the untenable and scientifically embarrassing results recovered in most, if not all current deep time genomic tests.

Avilla LS and Mothé D 2021. Out of Africa: A New Afrotheria Lineage Rises From Extinct South American Mammals. Frontiers in Ecology and Evolution 9:654302.
doi: https://doi.org/10.3389/fevo.2021.654302
Ezcurra MD and Agnolin F 2012. A New Global Palaeobiogeographical Model for the Late Mesozoic and Early Tertiary. Systematic Biology 61(4):553–566.


Bird palates compared

Earlier we looked at updates (= housekeeping) to the bird clade
of the large reptile tree (LRT, 1880+ taxa). Today a selection of bird palates in the LRT are presented in phylogenetic order. Not exactly cherry-picked, those not presented here either do not show the palate, or the palate is obscured along the rim by mandibles in ventral view.

birds are divided into Palaeognathae and Neognathae according to their palates. Palaeognathae include the ratites, “more primitive and reptilian than that in other birds.” According to Wikipedia, “McDowell (1948) asserted that the similarities in the palate anatomy of paleognathes might actually be neoteny, or retained embryonic features. He noted that there were other feature of the skull, such as the retention of sutures into adulthood, that were like those of juvenile birds. Parker (1864) reported the similarities of the palates of the tinamous and ratites, but Huxley (1867) is more widely credited with this insight.”

Cracraft (1974) defined Palaeognathae with five traits
and Feduccia (1980, 1996) defined Neognatha with three traits. In the LRT clades are based on a last common ancestor, not with traits. The short list of traits used by these authors too often converge elsewhere and both precede phylogenetic analysis software.

FIgure 1. Basal bird palates compared. Various sources. DGS colors added here. Based on sister taxa, the premaxilla likely extended back to the medial fenestra in the Pseudocrypturus and Rhynchotus drawings above.

In general, what do we see in basal birds?
in the Solnhofen bird, Jurapteryx (formerly Archaeopteryx, Fig. 1), the premaxillae are large, forming a palate, the maxillae are slender and lateral, the ectopterygoids are comma-shaped, and the rest are struts. That’s our starting point for today’s discussion.

In the basal crown bird and kiwi cousin, Pseudocrypturus
(Fig. 1), enlarging maxillae take the place of shrinking premaxillae. Teeth and ectopterygoids disappear.

In higher basal crown birds, like the ratites
(Fig. 1), the palatines become larger and more robust. Struthio, the ostrich opens up the palate with fenestrae, but other ratites, like Casurarius, don’t.

(Fig. 1) and its clade of Cretaceous birds redevelop a full arcade of teeth from Megapodius-like toothless ancestors. Premaxillary teeth are secondarily lost in Ichthyornis and Hesperornis.

In all the more derived extant and extinct birds
(Figs. 1–3) the palatal bones, premaxillae and maxillae assume such a wide variation in morphology based on diet and ancestry that they are best viewed and understood individually and comparatively by the reader, rather than described in detail with patterns and exceptions reported to potential excess in the text.

Figure 2. Palates of crown birds, terresrial clade. Various sources. DGS colors added here.

The greatest variation in palatal patterns
occurs with the relative amount of premaxillae to maxillae in palatal view. In some taxa the premaxillae dominate. In others the maxillae dominate. In a few the premaxillae underlie and laminate to the maxillae. The palatines and pterygoids do not change much. The vomers often disappear, but the large flightless taxa, Aptornis and Dinornis have large vomers. This is likely due to enbryonic retention via neotony (i.e. flightless wings also produce pre-loss vomers).

Figure 3. Palates of crown birds, aquatic clade. Various sources. DGS colors added here.

Adding colors to bones
to ease and speed identification is a method (aka Digital Graphic Segregation, or DGS) first offered in 2003 (e.g. the Jeholopterus skull that appears in the blog masthead above). The DGS method appears to be gaining wide acceptance (or wide convergence) nearly twenty years later for obvious beneficial reasons, whether applied to published µCT scans, traditional photographs or drawings. In the meantime, the switch from expensive 4-color printing press publication to no-cost RGB online publication has no doubt encouraged this transition, though still not universally and still not color-standardized.

Cracraft J 1974. Phylogeny and evolution of the ratite birds. IBIS 1116(4): 494–521.
Feduccia A 1980. The age of birds. Harvard University Press, Cambridge (Massachusetts) and London 196pp.
Feduccia A 1996. The origin and evolution of birds. New Haven, CT Yale University Press, 420pp.
Huxley TH 1867. On the classification of birds: and on the taxonomic value of the modifications of certain of the cranial bones observable in that class. Proceedings of the Zoological Society of London 1867:415–472, 36 figure.
McDowell S 1948. The bony palate of birds. Part 1. The Auk 65(4):520–549.
Parker WK 1864. V. On the Osteology of Gallinaceous birds and tinamous. The transactions of the Zoological Society of London 5(3):149–241.


Syntomiprosopus: might be a Triassic turtle with teeth

With only a short, broken, Late Triassic mandible to work with
several options remain open. Let’s seek maximum parsimony in homology here by testing a wide gamut of possible matching mandibles, leaving open all possibilities for convergence and novelty.

Heckert et al. 2021 described
four incomplete mandibles (Fig. 1) from “the Placerias/Downs’ Quarry complex in eastern Arizona, USA… a new short-faced archosauriform, Syntomiprosopus sucherorum gen. et sp. nov., that expands that diversity with a morphology unique among Late Triassic archosauriforms.”

Not too many Late Triassic archosauriforms have such a short face, as the authors duly note.

Figure 1. Syntomiprosopus is only known from mandible fragments here matched to several taxa. The best match may be to the transitional pareiasaur to meolanid turtle, Elginia, a clade not considered by Heckert et al. 2021.
Figure 1b. Frames from above not animated.

Heckert et al. 2021 continue:
“The most distinctive feature of Syntomiprosopus gen. nov. is its anteroposteriorly short, robust mandible with 3–4 anterior, a larger caniniform, and 1–3 “postcanine” alveoli. The size and shape of the alveoli and the preserved tips of replacement teeth preclude assignment to any
taxon known only from teeth.

The tooth placement pattern is certainly distinctive… something a little different.

“Additional autapomorphies of S. sucherorum gen. et sp. nov. include a large fossa associated with the mandibular fenestra, an interdigitating suture of the surangular with the dentary, fine texture ornamenting the medial surface of the splenial, and a surangular ridge that completes a 90° arc.

Can these traits also be found in pareiasaurs and basal turtles? See figure 1 for comparisons. The surangular ridge that completes a 90º arc is especially intriguing due to its rarity in tetrapods. None of the closest tested taxa on the toothed pareiasaur side of the transition preserve posterior mandible elements. However, if we extend our look to the pareiasaur, Anthodon, such a trait, though small, was figured earlier by Lee 1997 (Fig. 2).

Figure 2. Anthodon in various views from Lee (1997). Note the posterior surangular has a 90º radiant with a small lateral lip, less exaggerated than the same surangular in Syntomiprosopus (Fig. 1).

Heckert et al. 2021 continue:
The external surfaces of the mandibles bear shallow, densely packed, irregular, fine pits and narrow, arcuate grooves. This combination of character states allows an archosauriform assignment; however, an associated and similarly sized braincase indicates that Syntomiprosopus n. gen. may represent previously unsampled disparity in early-diverging crocodylomorphs.”

Autapomorphies in one clade are often synapomorphies in another omitted clade. Turtles and pareiasaurs are not mentioned in the test. Neither are Elginia or Meiolania, taxa transitional between the pareiasaur, Bunostegos, and hard-shelled turtles, like Proganochelys in the large reptile tree.

Mandibular fenestra?
Heckert et al. identified one part of the split between the dentary and post-dentary elements as the mandibular fenestra (Fig. 1). Maybe so, but that fenestra is only part of a larger break, so maybe not.

Heckert et al. report,
“Syntomiprosopus n. gen. may represent previously unsampled disparity in early-diverging crocodylomorphs.”

Or it could represent a late-surviving pareiasaur transitional to turtles. Let’s keep working on this taxon and keep all options open.

Heckert AB et al. (5 co-authors) 2021. A new short-faced archosauriform from the Upper Triassic Placerias/Downs’ quarry complex, Arizona, USA, expands the morphological diversity of the Triassic archosauriform radiation. The Science of Nature 108:32.
Lee MSY 1997. Pareiasaur phylogeny and the origin of turtles. Zoological Journal of the Linnean Society 120: 197-280.

When a simple torsion fracture turned an Early Cretaceous bird into bizarre bat-wing dinosaur

Today, another graphic of Yi qi
(this time with animation, Figs. 1, 2) of an earlier (2015) hypothesis that showed no styliform element in Yi qi (and Ambopteryx), only a broken ulna (Fig. 3).

The bat-wing bird hypothesis
promoted by Dececchi et al. 2020 and prior authors since 2015 (see below) is a myth. A misinterpretation. A false tradition that will not die. Here I am trying to put yet another nail in a coffin that will not close six years after the first nail.

Perhaps the most embarrassing aspect of this scandal
is the large number of professors, grad students, assorted paleontologists (see below) and paleoartists (see below) who gladly accepted a 2015 reconstruction of Yi qi that not only added a completely new long bone to the arm (the so-called ‘styliform’), but turned a perfectly good Early Cretaceous bird into a bat-wing dinosaur with flight membranes instead of feathers.

Completely forgotten
was the traditional dictum, “exceptional claims require exceptions evidence”... or in this case, just any meager evidence at all. No one else, using the most advanced imaging techniques (Fig. 1, Dececchi et al. 2020) and first-hand observation, were able to see that poor Yi qi simply suffered from a broken arm.

As an example,
five years after the broken arm had been noted online, Dececchi et al. 2020 were still trying to figure out how Yi qi would have flown with bat-wings and a fourth long arm bone. Highly regarded referees and editors, the traditional gate keepers of good data and good science, kept approving this nonsense in published works (see below).

Figure 1. Animation of Yi qi showing how the ulna rotated, splintered and gave everyone else the illusion of a new long bone in the forelimb, which led to the bat-wing myth. Illustrations and photos from Dececchi et al. 2020.

A closer view:

Figure 2. Closeup of figure 1. Illustrations and photos from Dececchi et al. 2020. Colors and animation added here.

From the Dececchi et al. 2020 summary:
“The bizarre scansoriopterygid theropods Yi and Ambopteryx had skin stretched
between elongate fingers that form a potential membranous wing.”

Not true. Given firsthand access to the fossil and laser stimulated fluoresence (LSF, Figs. 1, 2) imaging, this team of PhDs was still unable to describe the torsion break in the ulna of Yi qi — even after showing the break itself! Instead, they accepted that this taxon had developed a completely unique long arm bone (the ‘styliform’) while, at the same time, essentially losing a major traditional long arm bone (the ulna) during a taphonomic wing flip of 180º. Funny, odd, embarrassing that the many authors that have so far examined and written about Yi qi never put these facts together, whether by creating a reconstruction or by careful first-hand observation or by using Occam’s razor. It just had a broken arm.

To this day
no one has produced a tetrapod of any kind that has four long robust bones in each forelimb proximal to the manus. Three is the number. It has always been so. Yi qi is no exception. A long pterosaur pteroid is a possible contender, but is always much more slender than the radius and ulna. Pterosaur, bird and bat fingers are not under consideration here because they extend beyond the wrist and are well-accounted for.

Yi qi was nothing more than this:
an ordinary Early Cretaceous bird suffering from a broken ulna and an undeserved reputation.

Figure 3. Yi qi tracing of the in situ specimen from 2015 using DGS method and bones rearranged to form a standing and flying Yi qi specimen. Note the lack of a styliform element, here identified as a displaced radius and ulna. This is an Early Cretaceous bird suffering from a broken ulna and an undeserved reputation.

Question for you, dear reader:
Should Xu et al. 2015 and Deceechi et al. 2020 retract their papers? Like the Oculudentavis scandal, their claims are demonstrably false. Were these examples of headline grabbing by paleontologists who should have know better? Or just a lot of scientists making an honest mistake? As always, decisions like this are up to you, whose opinions ultimately create the consensus.

In theory, facts should override opinions.
However in practice, sometimes (in this case, since 2015) sensational opinions override facts.

Dececchi TA, et al. (8 co-authors) 2020.
Aerodynamics show membrane-winged theropods were a poor glidiing dead-end. iScience 2020 https://doi.org/10.1016/j.isci.2020.101574
Wang M, O’Connor JK.; Xu X and Zhou Z 2019. A new Jurassic scansoriopterygid and the loss of membranous wings in theropod dinosaurs. Nature 569: 256–259. doi:10.1038/s41586-019-1137-z
Xu X, Zheng X-T, Sullivan C, Wang X-L, Xing l, Wang Y, Zhang X-M, O’Connor JK, Zhang F-C and Pan Y-H 2015. A bizarre Jurassic maniraptoran theropod with preserved evidence of membranous wings. Nature (advance online publication)


Paleoartist John Conway made an illustration of Yi qi with bat wings you can see here.

Paeloartist Emily Willoughby made an illustration of Yi qi with bat wings you can see here.

Darren Naish blogging for Scientific Americanin 2015 wrote,
“It probably looked more like a bat-winged parrot.”

Darren Naish also reported, “Exactly such a creature was predicted by my colleague Andrea Cau (and illustrated by excellent palaeoartist Lukas Pankarin) way back in October 2008 after the publication of Epidexipteryx* (Zhang et al. 2008).

Naish continues, “Could Yi qi‘s styliform elements actually be battle spines or something? Xu et al. (2015, supplementary information) state that “we are aware of no case in which a long, unjointed bony or cartilaginous rod extending from a limb joint has evolved in any vertebrate without being associated with an aerodynamic membrane, and plausible alternative functions for such a structure are difficult to conceive”.

“It’s difficult to pinpoint where the mistakes have been made, but my suspicion is that Yi qi would have looked more like other maniraptorans, and less dragony overall.

Seemingly every explanation was offered, but the simplest explanation, the one you can see: a broken ulna.

Naish 2015 continues,
“It shouldn’t be lost on you that Yi qi and other scansoriopterygids look to have been experimenting with flight and climbing, despite being well outside the bird clade itself.”

In the LRT, Yi qi and other scansoriopterygids are birds derived from Solnhhofen birds, members of the bird clade itself.

Yi qi was made a plastic toy featured online here.
Googling “Yi qi – images” will bring you here to a panoply of images.

Some publicity at the time:

CM 81532 is not Anthracodromeus and not a reptile

Mann et al. 2021 brought us
CM 81352 (Figs. 1–3) a small Late Carboniferous skeleton they considered specimen #3 of the scansorial (= tree-climbing) basal archosauromorph reptile, Anthracodromeus (Fig. 1).

After testing
in the large reptile tree (LRT, 1880+ taxa) CM 81352 nests not with Anthracodromeus, but with other taxa that have a short, robust dorsal ribs, a single sacral rib and only four fingers with fewer phalanges among basal Reptilomorpha: Platyrhinops, Amphibamus, Balanerpeton and Caerorhachis (Fig. 1). Mann et al. did not mention any of these taxa in their text. So, taxon exclusion is once again a problem here.

Another problem:
Mann et al. added a finger that was not present. They wrote: “Only four digits are
preserved on either side, the fifth digit on the right manus and the fourth digit on the left manus likely became dissociated post-mortem.”

Or there were just four fingers. Evidence is evidence. And that means CM 81352 was not a basal reptile. Five (or more) fingers show up in the LRT several times from a basal tetrapod number of just four fingers.

Figure 1. CM 81532 to scale with basal reptilomorph sister taxa in the LRT along with the unrelated, but coeval reptile, Anthracodromeus

From the abstract:
“A new skeleton of the exceedingly rare, late Carboniferous eureptile Anthracodromeus longipes (Fig. 1, originally Sauropleura Cope1875, Carroll and Baird 1972), reveals the presence of a reduced phalangeal count in the manus and pedes and uniquely recurved claws.

No other basal reptiles related to Anthracodromeus had a reduced phalangeal count.

“With these data, we quantitatively evaluate the locomotor ecology of Anthracodromeus using morphometric analyses of the phalangeal proportions, claw curvature, and claw shape.”

What these authors need is an inclusive phylogenetic analysis BEFORE conducting a morphometric analysis. Taxon exclusion is the number one problem in paleo today.

“Our findings indicate that the anatomy of Anthracodromeus likely facilitated scansorial clinging to some degree via recurved claws and increased surface area of the large manus and pes.”

The reptilomorph Platyrhinops (Fig. 1) has similar manual and pedal proportions. By contrast Anthracodromeus has a larger manus relative to the pes and related reptiles have five fingers with a complete phalangeal count according to phylogenetic bracketing.

“This suggests that Anthracodromeus was among the earliest amniotes to show climbing abilities, pushing back the origins of scansoriality by at least 17 million years.”

The reptilomorphs are also Late Carboniferous, but much more primitive indicating an even earlier origin of scansoriality.

It further suggests that scansoriality arose soon after the origin of amniotes, allowing them to exploit a wide range of novel terrestrial niches.”

The LRT indicates that scansoriality arose prior to the origin of amniotes.

Figure 2. CM 81532 in situ. Second frame is tracing from Mann et al. with colors added here after a second tracing. See figure 3 for manus and pes details.

Mann et al. report,
“While once considered part of the paraphyletic Protorothyrididae, Anthracodromeus is
currently considered a basal eureptile with close relationships to the similarly-aged Cephalerpeton and later Early Permian Protorothyris.”

No to the former, yes to the latter according to the LRT, which divides reptiles at their genesis into Lepidosauromorpha and Archosauromorpha, not Eureptilia and Parareptilia. Mann et al. performed several morphometric analyses, but not a phylogenetic analysis.

Figure 3. CM 81532 manus and pes as originally interpreted and reinterpreted here. Mann et al. added a finger and interpreted the pes differently.

The CM 81532 manus has only four fingers
(Fig. 3), but Mann et al. imagined, proposed and needed a fifth finger to suite their preconception, their hope, their wish. This is science. Don’t wish for something that clearly is not there. Look elsewhere in the family tree to find taxa with four fingers and start your search there.

In like fashion,
the authors interpreted the pes as complete, but with fewer phalanges on several digits. Here (Fig. 3) the pes has the standard phalangeal count for reptiles and reptilomorphs. Missing distal phalanges for digits 4 and 5 are beneath other bones (digit 4) and just barely visible (digit 5).

The sacral count
Mann et al. considered the first two caudal ribs to be sacrals and the ilium to be unusually shaped and much taller than in any other taxon. They needed two sacrals if their specimen was going to be a reptile. Here the single sacral vertebra has no attached sacral rib (Fig. 2). The missing (= detached) sacral rib is now found in situ at the top of the now shorter ilium (Fig. 2), displaced during taphonomy.

Short, robust ribs
The basal archosauromorph reptile, Anthracoderomeus, has slender curved ribs that encircle the torso. By contrast, the CM 81532 specimen has shorter, straighter, more robust ribs, as in related reptilomorphs, like Platyrhinops (Fig. 1).

Among reptilomorphs, CM 81532 has the longest fingers and toes, all armed with sharp claws. Climbing seems to be this taxon’s special niche, convergent with Anthracodromeus, homologous with Platyrhinops (Fig. 1).

By the way,
since CM 81532 is not Anthracodromeus, anymore, nor a eureptile, it needs a new generic and specific name. Any suggestions?

Mann A, Dudgeon TW, Henrici AC, Berman DS and Pierce SE 2021. Digit and ungual morphology suggest adaptations for scansoriality in the late Carboniferous eureptile Anthracodromeus longipes. doi: 10.3389/feart.2021.675337